Designing a water runnel feature

[Daniel C. Miller]

Hello all,

I’m currently working on a park project in which one of the features will be a 150′ plus linear channel through which a set volume discharge of water will run through periodically. The runnel will be approximately one foot wide and 4-6″ deep at its upstream end. The runnel will be covered by decorative grate in some sections and exposed (open) in other sections. The concept is that park visitors will be able to follow (visually and audibly) this discharge as it travels down the runnel. 

My knowledge of fluid dynamics is limited, so I am trying to gather some information to help me accomplish the concept. I would like to determine the quantity of water and the rate of flow (how much water to release for how long) in order to achieve a recognizable volume throughout the length of the runnel. As part of the channel will be covered with a grate I would like to provide some audible turbulence within those sections. I assume that the volume and inclusion of “obstacles” within the channel would help achieve this.

I would welcome any information (direct experience or reference material suggestions) that anyone could provide to aide me in developing the design.

I look forward to and thank you for any information you may have.

Thank you,

Dan Miller

[Wes Arola, RLA]

What is the difference in grade from the top of the runnel to the bottom? I will ask that question for all, and offer the suggestion that the larger grade difference the higher the ability to generate audible turbulence via gravity. Materials will play a large roll in noises and the restriction of water and letting it “build up” and have the weight of the water generate pressure to ‘squirt or spray from and opening in the base of a reservoir could be an idea. It sounds like you would like to only pump or circulate the water from the bottom of the runnel back to the top. Take advantage of gravity and grade change to accomplish this noise and interest as to avoid additional pumping for fountains or spray heads. Be aware that any splashing or free falling of water will allow for evaporation.

good luck!

[Wes Arola, RLA]

also keep in mine if you plan to get a sheet of water to flow across or fall from a surface the surface must be perfectly level. keep in mine the materials and constructibility.

[Thomas J. Johnson]

OK, your question brought out the geek in me. Plus, being unemployed I need to keep my skills sharp… Here’s what I came up with…

150′ x 1′ x 3″ = 37.5 cubic feet of water to fill the runnel 3 inches deep.

1 cubic foot = 7.4805 gallons

37.5 x 7.4805 = 280.5 gallons to fill the runnel 3 inches deep.

This is where it gets a little funky… According to the Manning Equation (With a 2% slope [3’/150′] and a roughness coefficient of .013 ‘concrete’), you would need to move 4.5 CFS or approx. 2,000 GPM or 120,000GPH! That’s a big pump! It looks like just the pump would cost about $8,000.

The Manning Equation

The Manning Equation for U.S. units is: Q = (1.49/n)A(R2/3)(S1/2), Where

Q = volumetric water flow rate passing through the stretch of channel, ft3/sec,

A = cross-sectional area of flow perpendicular to the flow direction, ft2,

S = bottom slope of channel, ft/ft (dimensionless),

n = Manning rougness coefficient (empirical constant), dimensionless,

R = hydraulic radius = A/P, where

A = cross-sectional area of flow as defined above,

P = wetted perimeter of cross-sectional flow area, ft.

The Manning Equation can be expressed in terms of flow velocity instead of flow rate. Using the equation, V = Q/A as a definition for average flow velocity, the Manning Equation becomes:

V = (1.49/n)(R2/3)(S1/2), with average flow velocity in ft/sec.

Note that the Manning Equation is an empirical, dimensional equation. With the constant equal to 1.49, all of the parameters must have the units given above.
The Manning Roughness Coefficient

The Manning roughness coefficient, n, is an experimentally determined constant. Its value depends upon the nature of the channel and its surface. Tables giving values of n for different man-made and natural channel types and surfaces are available in many textbooks, handbooks and on-line. Here are a few typical values for n:

Brick – n = 0.015, new cast-iron – n = 0.012, concrete – n – 0.011 to 0.015, corrugated metal – n = 0.022

Read more: http://www.brighthub.com/engineering/civil/articles/52905.aspx#ixzz…

So that’s my quick stab at it…assuming a smooth consistent surface. Of course as you add texture that slows things down… You could make it 1-1.5% and have small drops into pools that would slow it down even more and add the sound / interest you’re looking for… If you only have 1.5″ of water that would make it a lot less expensive also (less water to move) If you can slow it down to .25CFS / 112GPM / 8,000GPH you can get a good pump for under $1,000. Otherwise, that could end up being one expensive runnel!

One other concern you might run into is the decorative grate. If you have a number of sections of grate, depending on color / contrast, you may have tripping hazard with people accidentally walking off of the grate into the runnel. You may want to consider carrying the paving material across the runnel in spots so that crossings are clear/distinctive. You could have the drops happen on the downhill side of the “bridge” causing people to pause on the bridge and look down into the pool… could be a neat effect.

[Conal gallagher]

Sounds a wonderful concept …The idea of a runnel is to provide the sound of water without actually seeing the water or if you do see it it’s on a occasional basis…. I have built several in gardens and invariably they have a lawn edge interspersed with planting. The variation in sound is provided by marginals in the runnel creating a backlog so you achieve natural heights and falls. I have never seen or used any man made blockages as nature quickly provided the necessary blockages. The necessary fall is actually minimal bearing in mind that the pump is doing most of the work.. And also remember that a runnel is based on an agricultural drainage ditch although in the romantic sense…So consider a Robert Hardy painting in an appropriate setting not much wider than a foot so that it is easy to step across. Depth is gernerally one to two feet max…Will dig up photo’s if you are interested..Water fall is constant and shallow. Best of luck…Sounds fun..

[mark foster]

We just did an almost totally flat “creek” feature, about 70 feet long with a similar effect, and we designed the flow for 1″ depth of continuous water to get the flowing /turbulant water effect you are talking about. And, you are correct about the obstacles helping make noise. We used river rounds 2-4″ diameter, flat across so the water “shallowed” and made a little noise.

Others have done the math so I will not go into it. Reputable pump manufacturers have engineers on staff who can give you what you need. If you give them your desired discharge rate (continuous flow needed at the top) , the running length & height from pump to discharge point, they can give you pump and pipe size recomendations. Remember to make clear that you need this performance at the discharge, and not at the pump–these will be two different values. Pump manufacturers also deal in different mesurements (gallons per hour–gph)

I tend to size my delivery systems 10-20% larger than what the calculations tell me, and I put a bleed-off valve near the pump on the discharge line, which can be manually adjusted to the flow required. Nothing is worse than having a pump installed that is a little too small–better to have more capacity, and be able to adjust it down if needed.

The most common mistake I have seen in this type of feature is not dealing with the “surge”–the amount of water it takes to “charge” or fill the system when turning it on, and dealing with this excess water when it is turned off. You will need to provide a water storage (tank,pond), and/or fresh water fill device to deal with this. This amount varies greatly based on slopes and design details.

You will also need to think about external pump vs. submersible– another discussion. I only use submersible pump (after years of misery with externals, and after the submersible got more reliable and efficient). Recomended submersible pump manufacturers: Tsurumi, Tidal Wave

[Daniel C. Miller]

Wow. First of all thank you for all your thoughtful and obviously informed input. Upon seeing your responses, there does appear to be some additional information it looks like I should provide:

– The runnel will be completely gravity fed (no pump anticipated) with slopes that may vary between 0.5% to 1%.
– The system will not be recirculating. Although we had initially explored a recirculating system, since a portion of the feature will be exposed for potential bodily contact, the County Health Department considers the feature a water play element and it becomes subject to some strict regulations. A recirculating feature would require restrooms with showers and infared and chemical treatment which the project has neither the space or budget.
– To reduce the water to drain “waste” we are designing the runnel to only contain water when activated by a “switch” in the vicinity of the water source. That switch would trigger the release of a set volume of water at a set flow rate just sufficient enough to produce a visible and audible “plug” of water that could be followed down the runnel. This is where the proper volume calculation would come in. Thank you Thomas for reminding me of some of those calculations and coefficients I have not used in a while.
-The grates will be ADA compliant and will occur wherever the runnel passes through a paved surface. The exposed runnel section occurs where the runnel passes through a lawn space. The plan is to use a light edging material (concrete or pavers) at the surface on either side of the runnel to provide a visual contrast against the darker green grass.

I’m sure there may be more I should explain, but I’ll leave it there for now. Please keep the suggestions, experience, comments coming. They are very helpful and appreciated.

[Thomas J. Johnson]

So, are you tying into a river upstream and then releasing the water back into the river down stream?

Is the “switch” / valve going to be electric? I.e. a “plug” of water is sent down the runnel every 10 or 15 min.

You could also design a system that acts like Japanese bamboo fountain (shishi odoshi) but on a massive scale. You can then just set the flow and leave it, no need for electrical on/off valves. The combination of flow / reservoir size / balance would determine the size and frequency of the plug… plus it would add another element of interest to the runnel.

[mark foster]

It is so interesting to see where people (myself included) go with these questions!

Dan– I think you want to know how much of a given volume of water will be used over a period of time? If so—-

This is a rule of thumb I use (based on my experience, and a little lower geometry/math): A continuous flow of 1 cubic inch of water will require about 250 gallons per hour, which at 7.5 gallons/cubic foot equals approximately 33 cubic feet per hour.

So for example: your runnel at 12″ wide x 1/2″ deep flow of water will take 1500 gph., or 200 cu. ft per hour. A 1″ deep water flow will take twice this, and so on.

p.s. Thomas– I enjoyed your refresher! It reminds me why I felt lucky to get C-‘s in site engineering classes back in the stone ages!

[Thomas J. Johnson]

lol Yeah, site engineering was my least favorite class and I too got a C-. I had a great professor and despite my grade, learned a lot but preferred to focus my attention on the “big idea” creative aspects of design classes at the time. Now I appreciate the fact that “how” things get built is very much a part of design. Plus it’s good for the old gray matter to work in ways it’s not usually inclined to… keeps the mind limber…

[Mark Sanford]

i love these discussions, what a great website!

[Author]

Posts